August 2019
Intermediate to advanced
342 pages
9h 35m
English
Despite the relative simplicity of the implementation of the Perceptron (simplicity here constitutes the strength of the algorithm, if compared to the accuracy of the predictions provided), it suffers from some important limitations. Being essentially a binary linear classifier, the Perceptron is able to offer accurate results only if the analyzed data can be linearly separable; that is, it is possible to identify a straight line (or a hyperplane, in case of multidimensional data) that completely bisects the data in the Cartesian plane:

If instead (and this is so in the majority of real cases) the analyzed data ...
Read now
Unlock full access